arXiv:2510.26038v1 Announce Type: cross Abstract: Knowledge distillation (KD) is an effective method for model compression and transferring knowledge between models. However, its effect on model’s robustness against spurious correlations that degrade performance on out-of-distribution data remains underexplored. This study investigates the effect of knowledge distillation on the transferability of “debiasing” capabilities from teacher models to […]
Through the Judge’s Eyes: Inferred Thinking Traces Improve Reliability of LLM Raters
arXiv:2510.25860v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as raters for evaluation tasks. However, their reliability is often limited for subjective tasks, when human judgments involve subtle reasoning beyond annotation labels. Thinking traces, the reasoning behind a judgment, are highly informative but challenging to collect and curate. We present a human-LLM […]
Multi-Agent Reinforcement Learning for Market Making: Competition without Collusion
arXiv:2510.25929v1 Announce Type: cross Abstract: Algorithmic collusion has emerged as a central question in AI: Will the interaction between different AI agents deployed in markets lead to collusion? More generally, understanding how emergent behavior, be it a cartel or market dominance from more advanced bots, affects the market overall is an important research question. We […]
Multiclass Local Calibration With the Jensen-Shannon Distance
arXiv:2510.26566v1 Announce Type: cross Abstract: Developing trustworthy Machine Learning (ML) models requires their predicted probabilities to be well-calibrated, meaning they should reflect true-class frequencies. Among calibration notions in multiclass classification, strong calibration is the most stringent, as it requires all predicted probabilities to be simultaneously calibrated across all classes. However, existing approaches to multiclass calibration […]
Faithful and Fast Influence Function via Advanced Sampling
arXiv:2510.26776v1 Announce Type: cross Abstract: How can we explain the influence of training data on black-box models? Influence functions (IFs) offer a post-hoc solution by utilizing gradients and Hessians. However, computing the Hessian for an entire dataset is resource-intensive, necessitating a feasible alternative. A common approach involves randomly sampling a small subset of the training […]
Latent Chain-of-Thought for Visual Reasoning
arXiv:2510.23925v2 Announce Type: replace Abstract: Chain-of-thought (CoT) reasoning is critical for improving the interpretability and reliability of Large Vision-Language Models (LVLMs). However, existing training algorithms such as SFT, PPO, and GRPO may not generalize well across unseen reasoning tasks and heavily rely on a biased reward model. To address this challenge, we reformulate reasoning in […]
UniSite: The First Cross-Structure Dataset and Learning Framework for End-to-End Ligand Binding Site Detection
arXiv:2506.03237v2 Announce Type: replace Abstract: The detection of ligand binding sites for proteins is a fundamental step in Structure-Based Drug Design. Despite notable advances in recent years, existing methods, datasets, and evaluation metrics are confronted with several key challenges: (1) current datasets and methods are centered on individual protein-ligand complexes and neglect that diverse binding […]
Pass@K Policy Optimization: Solving Harder Reinforcement Learning Problems
arXiv:2505.15201v3 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) algorithms sample multiple n>1 solution attempts for each problem and reward them independently. This optimizes for pass@1 performance and prioritizes the strength of isolated samples at the expense of the diversity and collective utility of sets of samples. This under-utilizes the sampling capacity, limiting exploration and eventual […]
Capillarity Reveals the Role of Capsid Geometry in HIV Nuclear Translocation
arXiv:2510.26357v1 Announce Type: cross Abstract: The protective capsid encasing the genetic material of Human Immunodeficiency Virus (HIV) has been shown to traverse the nuclear pore complex (NPC) intact, despite exceeding the passive diffusion threshold by over three orders of magnitude. This remarkable feat is attributed to the properties of the capsid surface, which confer solubility […]
The Information-Theoretic Imperative: Compression and the Epistemic Foundations of Intelligence
arXiv:2510.25883v1 Announce Type: new Abstract: Existing frameworks converge on the centrality of compression to intelligence but leave underspecified why this process enforces the discovery of causal structure rather than superficial statistical patterns. We introduce a two-level framework to address this gap. The Information-Theoretic Imperative (ITI) establishes that any system persisting in uncertain environments must minimize […]